Professional Documents
Culture Documents
Metric Examples PDF
Metric Examples PDF
The following sections (B1 – B5) show the list of success indicators and their corresponding
metrics. The list is taken from the output of a systematic review on evaluating SPI initiatives.
1
B2: Project measures
Success Indicator Commonly Used Metrics
i) Defect density, DD = No. defects / size
- No. defects according to development phase
- Size in LOC, FP, etc
ii) Defect distribution (%) = No. of <Phase> category defects / Total No.
of defects
<Phase> := Analysis, Requirement, Design, Code, Document, Others
iii) Yield (% of defects found and fixed before compilation and testing)
iv) Fault-slip through (FST) = Total no. of faults in late verification /
total no. of fault in the project
Defects v) Number of defects found per <Phase>
<Phase> := Analysis, Requirement, Design, Code Review, Unit Test,
Others
vi) No. of injected defect / <Phase>
<Phase> := Analysis, Requirement, Design, Code Review, Unit Test,
Others
vii) Defect Injection Rate (DIR) of requirements, design, coding and
testing activities = Defects injected at the activity / all defects of the
project
i) Cost / Size
- Size in LOC, FP, etc
ii) Cost of rework
iii) Rework effort in person-hour, staff-hour, etc
iv) Effort * cost per staff-hour
Cost v) Person-month, person-hour, staff-hour, etc
vi) Rework index = % project effort spent in rework
vii) Avoidable Fault Cost (AFC) = Determined by FST and the average
estimated cost of finding a fault in a specific phase of the software
life-cycle
i) Project cycle time (days, months, etc)
ii) Cycle time normalized by size
iii) Project duration = Time of gating review that decides to release the
project to the customer or market - time of gating review which kicks
Schedule off the project
iv) Relative integration time = (Integration time / overall project time) %,
time is calculated in months
v) X factor = The amount of calendar time for the baseline project to
develop a product divided by the cycle time for the new project
i) Size / effort
- Size in LOC, FP, etc
- Effort in person-month, person-hour, hour, etc
ii) Effort / size
Productivity - Effort in person-month, person-hour, hour, etc
- Size in LOC, FP, etc
iii) Number of tasks solved per time unit
iv) Actual productivity = No. of written code / person hours
v) Apparent productivity = (No. written code + no. of reused code)
2
/person-hours
vi) Rate of delivery = No. of projects delivered per year
vii) Productivity = (Number of features in project / total effort in minutes)
* (2,400 minutes / week)
Schedule:
i) Schedule variance = (Actual duration- planned duration) / planned
duration
ii) No. of days ahead of schedule/behind schedule of release
iii) Schedule performance index (SPI) = Earned value / planned value
iv) Schedule Estimation Accuracy (SEA) = Actual project duration /
estimated project duration
v) Schedule overruns = 100 * (actual value - estimated value) / estimated
value
Effort:
i) Effort variance = (Actual effort - planned effort) / planned effort
ii) Project Visibility Index (PVI) = Planned person-months / actual
person-months
iii) Effort Estimation Accuracy (EEA) = Actual project effort / estimated
project effort
Estimation iv) 100 * (actual effort - estimated effort) / estimated effort
Cost:
i) Cost performance index (CPI) = Earned value / actual cost
ii) Variance in the planned vs. actual cost for projects
Size:
i) Size variance = (Planned size - actual size) / planned size
ii) Error rate of size estimate = (LOC estimated – real LOC developed) /
real LOC developed.
iii) 100 * (actual LOC - estimated LOC) / estimated LOC
Others:
i) Productivity variance = Estimated vs. actual LOC / hour
ii) Cumulative Earned Value vs. Cumulative Planned Value
iii) Actual no. of defects removed per phase vs. estimated no. of defects
removed per phase
3
Reliability:
i) Failure Rate (FR) = Number of failures / execution time
ii) Mean Time Between Failure (MTBF) = (Downtime –uptime) / no. of
failures
iii) Mean Time To Failure (MTTF)
iv) Defect density by Configuration Mgmt build (DDb)
Maintainability:
i) McCabe's Cyclomatic complexity
ii) Maintainability product (MP) = Change scope * effort * 100
Note: The Change scope measure examines what percentage of the
program's components were added or modified to implement a
change.
iii) LOC/subroutine
iv) Stability index = Percent of the code that has changed in a week,
number of developers that made changes in a week, average number
of LOC per developer, average time spent by a developer on the
project
Reusability:
i) Reuse percent = Part of baseline against which projects are evaluated
ii) Reusability = Uses of delivery package = Amount of project specific
development done in the context of the delivery package
i) Cost = Effort * cost per staff-hour (measured in base cost, addition
cost for tool, saving through insertion of tool for varying project size
calculated in features and for two aspect regular process to advance
process and advance process to complete process)
ii) Cost = Training Cost + implementation cost
iii) Cost = cost of conformance + cost of non-conformance; cost non-
conf. is direct and indirect cost with reworking a product / service.
Conformance cost is appraisal and prevention cost. Appraisal is
evaluating the product for faults. Prevention is action to reduce or
eliminate rework.
Cost iv) Support cost = % of total effort for customer support
v) Relative cost of software testing = (Testing effort / overall project
effort)%, effort is calculated in man hours
vi) Investment cost = The quantity of the project-related investment costs
(training and initial tool usage problems) was obtained through a
questionnaire where each developer estimated how much of the
reported time they put on different activities (where training and
initial tool usage problems where two of those)
vii) Development support effort = Number of days effort expended by the
development team on defect investigation and resolution during the
system test period.
i) Time to deliver (days, months, etc)
ii) Time-to-market = The elapsed time between system conception and
deployment
Time-to-Market iii) (Delivery date(days) – Contract Date (Days) / 30)
iv) No. of validation cycles (code/test/re-code) before customers'
requirements are met
4
B4: Organization measures
Success Indicator Commonly Used Metric
Note: The above are not really metrics and the usual way of measuring the
employee communication is through survey / questionnaires.
5
B5: External measures
Success Indicator Commonly Used Metric
i) Operation productivity
ii) Product quality
iii) Process efficiency
iv) Product sales revenue
v) Net profit
vi) Improved in business operations*
vii) Improved in employee satisfaction*
Customer
Note: The above are some general success indicators that can be used to
measure the impact of SPI on the customer side. In fact, all the other
success indicators listed in the Process, Project, Product and Organization
can be used depending on how and where the product is being used in the
customer’s organization. The ones marked with * not really metrics and
the can be measured using survey / questionnaires.
Note: The above are some general areas that we can measure the impact of
the SPI to the society. The exact metrics are depending on the specific
product that is directly / indirectly used by the people.